|
on Forecasting |
| By: | Boriss Siliverstovs (KOF Swiss Economic Institute, ETH Zurich, Switzerland); Konstantin A. Kholodilin (DIW Berlin, Germany) |
| Abstract: | This study utilizes the dynamic factor model of Giannone et al. (2008) in order to make now-/forecasts of GDP quarter-on-quarter growth rates in Switzerland. It also assesses the informational content of macroeconomic data releases for forecasting of the Swiss GDP. We find that the factor model offers a substantial improvement in forecast accuracy of GDP growth rates compared to a benchmark naive constant-growth model at all forecast horizons and at all data vintages. The largest forecast accuracy is achieved when GDP nowcasts for an actual quarter are made about three months ahead of the official data release. We also document that both business tendency surveys as well as stock market indices possess the largest informational content for GDP forecasting although their ranking depends on the underlying transformation of monthly indicators from which the common factors are extracted. |
| Keywords: | Business tendency surveys, Forecasting, Nowcasting, Real-time data, Dynamic factor model |
| JEL: | C53 E37 |
| Date: | 2010–01 |
| URL: | http://d.repec.org/n?u=RePEc:kof:wpskof:10-251&r=for |
| By: | Daniel PREVE (School of Economics, Singapore Management University); Anders ERIKSSON (Department of Information Science/Statistics, University of Uppsala); Jun YU (School of Economics, Singapore Management University) |
| Abstract: | This paper introduces a parsimonious and yet flexible nonnegative semiparametric model to forecast financial volatility. The new model extends the linear nonnegative autoregressive model of Barndorff-Nielsen & Shephard (2001) and Nielsen & Shephard (2003) by way of a power transformation. It is semiparametric in the sense that the dependency structure and distributional form of its error component are left unspecified. The statistical properties of the model are discussed and a novel estimation method is proposed. Simulation studies validate the new estimation method and suggest that it works reasonably well in finite samples. The out-of-sample performance of the proposed model is evaluated against a number of standard methods, using data on S&P 500 monthly realized volatilities. The competing models include the exponential smoothing method, a linear AR(1) model, a log-linear AR(1) model, and two long-memory ARFIMA models. Various loss functions are utilized to evaluate the predictive accuracy of the alternative methods. It is found that the new model generally produces highly competitive forecasts. |
| Keywords: | Autoregression, nonlinear/non-Gaussian time series, realized volatility, semiparametric model, volatility forecast. |
| Date: | 2009–11 |
| URL: | http://d.repec.org/n?u=RePEc:siu:wpaper:22-2009&r=for |
| By: | Buss, Ginters |
| Abstract: | The paper compares one-period ahead forecasting performance of linear vector-autoregressive (VAR) models and single-equation Markov-switching (MS) models for two cases: when leading information is available and when it is not. The results show that single-equation MS models tend to perform slightly better than linear VAR models when no leading information is available. However, if reliable leading information is available, single-equation MS models tend to give somewhat less precise forecasts than linear VAR models. |
| Keywords: | Markov-switching; VAR; forecasting; leading information |
| JEL: | C32 C13 C51 C53 C52 C22 |
| Date: | 2010–02–14 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:20688&r=for |
| By: | Rangan Gupta (Department of Economics, University of Pretoria); Alan Kabundi (Department of Economics and Econometrics, University of Johannesburg); Stephen M. Miller (Department of Economics, University of Nevada, Las Vegas) |
| Abstract: | We employ a 10-variable dynamic structural general equilibrium model to forecast the US real house price index as well as its turning point in 2006:Q2. We also examine various Bayesian and classical time-series models in our forecasting exercise to compare to the dynamic stochastic general equilibrium model, estimated using Bayesian methods. In addition to standard vector-autoregressive and Bayesian vector autoregressive models, we also include the information content of either 10 or 120 quarterly series in some models to capture the influence of fundamentals. We consider two approaches for including information from large data sets – extracting common factors (principle components) in a Factor-Augmented Vector Autoregressive or Factor-Augmented Bayesian Vector Autoregressive models or Bayesian shrinkage in a large-scale Bayesian Vector Autoregressive models. We compare the out-of-sample forecast performance of the alternative models, using the average root mean squared error for the forecasts. We find that the small-scale Bayesian-shrinkage model (10 variables) outperforms the other models, including the large-scale Bayesian-shrinkage model (120 variables). Finally, we use each model to forecast the turning point in 2006:Q2, using the estimated model through 2005:Q2. Only the dynamic stochastic general equilibrium model actually forecasts a turning point with any accuracy, suggesting that attention to developing forward-looking microfounded dynamic stochastic general equilibrium models of the housing market, over and above fundamentals, proves crucial in forecasting turning points. |
| Keywords: | compensating variation, nonlinear income effects, discrete choice |
| JEL: | Q51 R21 |
| Date: | 2009–12 |
| URL: | http://d.repec.org/n?u=RePEc:nlv:wpaper:1001&r=for |
| By: | Meyler, Aidan; Rubene, Ieva |
| Abstract: | This document provides a summary of the aggregate results of a special questionnaire which was sent to the participants in the ECB Survey of Professional Forecasters (SPF) in autumn 2008, in the context of the ten-year anniversary of the SPF’s launch in January 1999. In summary, the results show that the SPF responses are quite timely and that the forecasts are based on heterogeneous assumptions that are predominantly generated in house. In addition, although both structural and time series models are widely used, judgement also plays a key role, in particular for the reported probability distributions. It is thus very important to consider the heterogeneity of the SPF forecasts when analysing and interpreting the results of the SPF. |
| Keywords: | SPF; Survey; Forecasts; Expectations; Formation |
| JEL: | D81 D84 C53 E17 C42 E37 |
| Date: | 2009–04 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:20751&r=for |